Dynamic geological model updating and prediction method based on tunnel face information

Abstract

The application provides a tunnel face information-based dynamic geological model updating and predicting method, which combines the tunnel face grid point data obtained in the tunnel excavation process, adopts Kriging interpolation and variogram fitting technology, and constructs and dynamically updates the geological model. The method comprises the following steps: collecting the face data, cleaning and standardizing the data, and extracting the tunnel starting end face and the ending end face; generating an initial geological model by using a three-dimensional variogram model and Kriging interpolation; integrating the newly added face data, dynamically adjusting the model parameters, and updating the geological distribution; and finally displaying the dynamically updated geological model through three-dimensional visualization technology. The application solves the problem that the traditional static model is difficult to adapt to the dynamic changes of geological conditions in the excavation process, can optimize the geological model in real time, provides accurate and reliable geological information support for tunnel construction, and significantly improves the construction efficiency and safety.

Description

Technical Field

[0001] This invention relates to a geological modeling and prediction technology for tunnel engineering, and in particular to a dynamic geological model updating and prediction method based on tunnel face information. Background Technology

[0002] During tunnel construction, dynamic changes in geological conditions have a significant impact on construction safety and design optimization. However, most current research on 3D geological modeling focuses on overall regional geological modeling based on geological survey information. These models are often static and struggle to dynamically adapt to the complex geological conditions encountered during tunnel excavation. For example, during tunnel excavation, the stratum distribution may change locally, lithological characteristics may abruptly change, and groundwater permeability conditions may also dynamically change. These factors directly affect construction efficiency and may lead to serious safety hazards.

[0003] Traditional geological modeling methods typically rely on initial exploration data, constructing static geological models using limited borehole data or cross-sectional measurements. This approach suffers from high uncertainty in prediction results, especially during tunnel excavation. Because it cannot dynamically integrate newly added geological data from the tunnel face, model updates lag, failing to reflect timely changes in actual geological conditions. This lag not only reduces the predictive accuracy of the geological model but may also lead to adjustments in construction plans lacking a scientific basis, significantly increasing construction risks and project costs. Therefore, there is an urgent need for a modeling method that can incorporate dynamic data and reflect newly discovered geological information in real time during tunnel face advancement, achieving more accurate geological predictions and providing scientific and reliable technical support for tunnel construction. Summary of the Invention

[0004] This invention provides a dynamic geological model updating and prediction method based on tunnel face information. By combining tunnel face data, Kriging interpolation modeling, and dynamic updates, it achieves high-precision prediction of the tunnel and surrounding geological areas, and utilizes 3D visualization technology to display the model results in real time. The advantage of this method lies in its ability to dynamically integrate newly added tunnel face data and optimize the geological model in real time, solving the problem that traditional static models cannot adapt to the dynamic changes in geological conditions during tunneling. Furthermore, it iterates and updates the prediction results as tunnel excavation progresses, providing more reliable geological information support for the construction process.

[0005] A dynamic geological model update and prediction method based on tunnel face information includes the following steps:

[0006] The first step is to import the initial face grid point data from the tunnel excavation process, specifically as follows:

[0007] Open the file storing the initial face data of the tunnel;

[0008] The file contains the three-dimensional coordinates (X, Y, Z) of the grid points and their geological attribute values ​​(D1, D2, ... D). n );

[0009] Since stratigraphic attributes are usually discrete categories (such as stratigraphic D1, D2, D3, etc.), and traditional kriging interpolation methods are mainly applicable to continuous data, the geological attribute values ​​of each grid point are represented by binary variables 0 and 1. That is, the value of each attribute is either 1 (indicating that the point belongs to the stratigraphic layer) or 0 (indicating that the point does not belong to the stratigraphic layer).

[0010] For example: if a point belongs to stratum D3, then set D1=0, D2=0, D3=1;

[0011] Check that the file format is correct (such as EXCEL format) to ensure that the data can be parsed correctly.

[0012] Extract initial data for the grid points at the working face:

[0013] Read the data of each grid point in the file, including coordinate information and attribute values;

[0014] Extract the grid points with the minimum Z value as the data for the tunnel's starting face;

[0015] The grid points with the maximum Z value are extracted as data for the tunnel termination face.

[0016] The second step is to preprocess the imported data to ensure data quality, specifically:

[0017] Clean the initial face data:

[0018] Check for duplicate grid points in the data and determine whether there are redundant points based on the coordinate information;

[0019] Remove duplicate points to ensure that each grid point appears only once in the dataset;

[0020] After data preprocessing, a mesh point dataset of the working face is generated. This dataset is used as input to directly fit the three-dimensional variogram function and construct the initial geological model.

[0021] The third step is to construct a preliminary geological model of the tunnel based on the initial data, specifically:

[0022] Based on the preprocessed dataset, calculate the maximum distance between grid points:

[0023] Calculate the Euclidean distance between every two grid points one by one, using the following formula:

[0024] Distance = [(X2-X1)] 2 +(Y2-Y1) 2 +(Z2-Z1)2 ] 1 / 2

[0025] Record the distance between all point pairs, determine the maximum distance between grid points, and then divide the variogram interval to ensure the accuracy of the interpolation model.

[0026] Estimate the semivariance:

[0027] Divide the distance between point pairs into several intervals;

[0028] Within each interval, calculate the average of the squared differences in attribute values ​​for corresponding point pairs, using the following formula:

[0029]

[0030] in

[0031] N is the number of point pairs.

[0032] Z i and Z j For point-to-point attribute values.

[0033] Fitting the initial three-dimensional variogram model:

[0034] During the fitting process, the error is calculated using the least squares method, and a suitable variogram function, such as an exponential model or a spherical model, is selected to fit the initial semivariogram curve.

[0035] The choice of the variogram model is based on the actual distribution characteristics of the semivariogram: when the semivariogram increases exponentially with increasing distance, the exponential model is chosen; when the semivariogram tends to plateau after increasing to a certain distance, the spherical model is chosen.

[0036] An exponential model can be used to fit the variogram, as shown in the following formula:

[0037]

[0038] in

[0039] C0 is the sill value, representing the degree of variability at extremely short distances (usually zero distance), reflecting measurement errors or irregularities at very small scales.

[0040] C represents the nugget value, indicating the maximum variation in geological properties over a wider range.

[0041] 'a' represents the range, indicating the distance the variogram reaches the nugget value C.

[0042] The fitting results provide a spatial correlation description of geological properties for Kriging interpolation.

[0043] Using the fitted variogram, the preprocessed face grid point data is input as conditional data into the Kriging interpolation model.

[0044] The interpolation process is based on the following formula:

[0045] in,

[0046] Z * (X) is the predicted value.

[0047] λ i As weight,

[0048] Z(X i ) represents the attribute value of a known data point.

[0049] Generate the distribution prediction results of the initial tunnel geological properties.

[0050] The fourth step is to import the newly added tunnel face data during the tunnel excavation process and update the geological model, specifically as follows:

[0051] Open the newly added face grid point data file:

[0052] Whenever the tunnel excavation has been completed to a certain distance, the three-dimensional coordinates and geological attribute values ​​of the newly added face grid points are recorded to ensure that the new data covers the construction area;

[0053] Check the data file to ensure that the format is consistent with the initial data to avoid data reading errors.

[0054] Integrate the new data into the original dataset:

[0055] Merge the newly added face grid point data with the original dataset to ensure the integrity of coordinate information and attribute values;

[0056] Remove duplicate grid points and re-standardize the data range to ensure consistency.

[0057] Calculate the semivariance and fitted variogram of the updated preprocessed data, and adjust the parameters of the Kriging interpolation model:

[0058] Recalculate the point-to-point distances based on the new data and update the semivariance values;

[0059] After updating the semivariance, the weight matrix of the Kriging interpolation model is refitted to predict the region affected by the new data:

[0060] Use the newly added face grid point data to input the updated Kriging interpolation model;

[0061] Based on the newly added data's three-dimensional coordinates (X, Y, Z) and corresponding geological attribute values;

[0062] Interpolate the newly added area point by point and output the predicted attribute value for each grid point.

[0063] Interpolate the grid points in the surrounding area to generate a distribution prediction of geological attributes, ensuring the integrity of geological information in the surrounding area;

[0064] Generate updated geological attribute distribution data, covering the affected area of ​​the newly added working face data.

[0065] The fifth step is to normalize the prediction results and generate probability values ​​for geological attributes, specifically:

[0066] Probability values ​​for generating geological attributes:

[0067] In the prediction model, the geological attribute value of each grid point is represented by binary variables 0 and 1. After performing Kriging interpolation on the attribute values ​​(D1, D2, D3) of each stratum, the interpolation result is a continuous value, representing the "probability" or "possibility" of the predicted point belonging to a certain stratum.

[0068] For example, if the interpolation results for a certain prediction point are D1 = 0.7, D2 = 0.2, and D3 = 0.1, it means that there is a 70% probability that the point belongs to stratum D1, a 20% probability that it belongs to stratum D2, and a 10% probability that it belongs to stratum D3.

[0069] After interpolation, the stratigraphic attribute values ​​at some points may not meet the probability distribution requirements (e.g., the sum may be greater than or less than 1). To ensure the comparability of different stratigraphic attribute values, the geological attribute values ​​of each grid point are normalized. After normalization, the attribute values ​​are standardized into probability values, with the numerical range limited to between 0 and 1, and the sum of all geological attribute values ​​is 1.

[0070] For each grid point's prediction results, the geological attribute values ​​are dynamically normalized using the following formula:

[0071]

[0072] P i Geological attribute values ​​for each grid point

[0073] Generate a probability list to determine the affiliation of grid points in different geological layers;

[0074] The correctness of the processing results is ensured by verifying whether the sum of the normalized probability values ​​of each grid point is 1.

[0075] The sixth step is to perform a three-dimensional visualization of the dynamically predicted geological model, specifically as follows:

[0076] Open the predicted data results file:

[0077] Read the 3D grid points and their corresponding geological attribute distribution data after the prediction is completed.

[0078] Constructing a 3D geometric point cloud:

[0079] Based on geometric mapping theory, software with similar functions, such as ParaView, is used to convert 3D mesh point data into point cloud models.

[0080] Assign an attribute value to each grid point as a color label for the point cloud.

[0081] Draw an attribute distribution diagram:

[0082] Draw point clouds to show the distribution of geological properties in and around the tunnel;

[0083] Set up a color mapping scheme to assign different colors to different geological attribute values;

[0084] Use a color mapping scheme to clarify the correspondence between the geological attribute values ​​and colors;

[0085] Add color bars to help identify the probability distribution of geological attributes in different areas. Attached Figure Description

[0086] Figure 1 This is a schematic diagram of the overall process of a dynamic geological model update and prediction method based on tunnel face information provided in an embodiment of the present invention;

[0087] Figure 2 This invention provides a method for updating and predicting dynamic geological models based on tunnel face information, illustrating the representation of tunnel face information.

[0088] Figure 3 A schematic diagram of the tunnel face grid points for a dynamic geological model update and prediction method based on tunnel face information provided in this embodiment of the invention;

[0089] Figure 4 A preliminary fitting semivariogram diagram of a dynamic geological model update and prediction method based on tunnel face information provided in an embodiment of the present invention;

[0090] Figure 5 A preliminary geological model rendering of a dynamic geological model update and prediction method based on tunnel face information provided in this embodiment of the invention;

[0091] Figure 6 A schematic diagram of the updated fitting semivariogram of a dynamic geological model update and prediction method based on tunnel face information provided in an embodiment of the present invention;

[0092] Figure 7A geological distribution probability map for a dynamic geological model update and prediction method based on tunnel face information provided in this embodiment of the invention;

[0093] Figure 8 This is a visualization of the dynamically updated geological model, provided by an embodiment of the present invention, based on a dynamic geological model updating and prediction method using tunnel face information. Detailed Implementation

[0094] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0095] See Figures 1-8 This invention provides a method for updating and predicting dynamic geological models based on tunnel face information, comprising the following steps:

[0096] To construct the geological model, grid point information was first extracted from the file storing tunnel face data. The extracted information included the three-dimensional coordinates (X, Y, Z) and geological attribute values ​​(D1, D2, D3) of each grid point. The geological attribute value of each grid point was represented by binary variables 0 and 1. If a grid point belonged entirely to a specific stratum (e.g., D1), D1 was set to 1, while the attribute values ​​for other strata (e.g., D2, D3) were 0. For example, if a point belonged to stratum D3, D1 = 0, D2 = 0, and D3 = 1. Figure 2 As shown.

[0097] After extraction, the grid point data undergoes preprocessing, during which duplicate points are removed to ensure data uniqueness. The cleaned data will be used for interpolation modeling and dynamic updates. Then, based on the Z-value, the starting and ending faces of the tunnel are extracted, corresponding to the sets of grid points with the minimum and maximum Z-values, respectively, providing boundary conditions for the initial model construction, such as... Figure 3 As shown.

[0098] After data preprocessing, the maximum distance between grid points is obtained by calculating the Euclidean distance. Based on this, the distance between point pairs is divided into several intervals, and the semivariogram value within each interval is calculated, representing the geological attribute value of the point pair. The semivariogram value increases rapidly with increasing distance, then tends to stabilize, but fluctuates at longer distances. Therefore, an exponential model is used to fit the semivariogram curve, such as... Figure 4As shown. After fitting, an initial geological model is generated using the Kriging interpolation method to predict the distribution of geological properties in the tunnel and surrounding area, such as... Figure 5 As shown.

[0099] Every 25m of tunnel excavation, newly added grid point data at the tunnel face are collected, extracting 3D coordinates (X, Y, Z) and geological attribute values ​​(D1, D2, D3). The new data is then integrated with the original dataset. After the new data enters the system, it undergoes cleaning and deduplication operations to ensure data accuracy.

[0100] The integrated data is used to recalculate the semivariance and adjust the weight matrix of the kriging interpolation model to reflect the spatial characteristics of the new data, such as... Figure 6 As shown. Subsequently, the updated Kriging interpolation model was used to re-predict the geological attribute distribution of the tunnel and surrounding area, generating the updated geological model probability distribution, as shown. Figure 7 As shown.

[0101] To visually demonstrate the prediction results of the geological model, the calculated grid point data was exported as a standardized Excel file and converted to CSV format before being imported into ParaView software for 3D visualization. A 3D point cloud model was generated using geometric mapping technology, mapping the coordinates (X, Y, Z) of each grid point into 3D space. Based on the predicted geological attribute values, a corresponding color value was assigned to each grid point, with the color intensity directly reflecting the differences in stratigraphic attribute values. In the visualization model, by updating the calculated data file, newly added face data can be loaded in real time, and the point cloud model can be incrementally adjusted to promptly reflect the impact of new data on the geological model. Simultaneously, ParaView's color mapping function was used to set multiple color gradient schemes, assigning distinct colors to different geological attribute values ​​and visually displaying the spatial distribution of geological attributes. Furthermore, color bars were added to clearly indicate the correspondence between geological attribute values ​​and colors, ensuring that construction personnel can quickly identify stratigraphic types and their attribute distribution.

[0102] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for updating and predicting dynamic geological models based on tunnel face information, characterized in that, Includes the following steps: The first step involves importing geological information from the tunnel face grid points during tunnel excavation to form an initial tunnel face dataset. The geological attributes of these grid points are discrete categorical geological attributes, represented by binary variables 0 and 1. The second step preprocesses the grid point information, including removing duplicate data. The third step constructs a geological model of the tunnel face based on the initial acquired data and uses Kriging interpolation, combined with a fitted variogram model, to calculate the preliminary geological attribute distribution of the tunnel and its surrounding area. The fourth step recalculates the point-to-point distance, semivariance, and variogram based on newly added grid point data during tunnel excavation. The Kriging interpolation model parameters are dynamically adjusted, and the geological attribute distribution of the tunnel and its surrounding area is re-predicted based on the dynamically updated model, further optimizing the model's accuracy. The fifth step normalizes the predicted geological attribute values, generating probability values ​​for various geological attributes to ensure consistency in the prediction results. The sixth step uses 3D point cloud rendering technology to visualize the dynamically updated geological model in real time, intuitively displaying the distribution and dynamic changes of the tunnel's geological attributes.

2. The dynamic geological model updating and prediction method based on tunnel face information according to claim 1, characterized in that: When importing grid point data from the tunnel face, this includes obtaining the three-dimensional coordinates (X, Y, Z) of the grid points and their corresponding geological attribute values ​​(D1, D2, ... D). n And store it in an Excel data file.

3. The dynamic geological model updating and prediction method based on tunnel face information according to claim 1, characterized in that: When preprocessing the grid point data at the working face, duplicate data points are removed to ensure data integrity and consistency.

4. The dynamic geological model updating and prediction method based on tunnel face information according to claim 1, characterized in that: Based on the preprocessed tunnel face data, the maximum distance between grid points is calculated, the semivariance value is estimated in intervals, and a preliminary three-dimensional variogram model is fitted. Based on the fitted preliminary variogram model, the geological properties of the tunnel and the surrounding area are predicted.

5. The dynamic geological model updating and prediction method based on tunnel face information as described in claim 1, characterized in that: Based on the newly added grid point data at the tunnel face during tunnel excavation, the parameters of the Kriging interpolation model are dynamically adjusted, including recalculating the point-to-point distance, semivariance value, and updated three-dimensional variogram curve. Based on the updated interpolation model, the geological property distribution of the tunnel area and its surrounding area is re-predicted.

6. The dynamic geological model updating and prediction method based on tunnel face information according to any one of claims 1-5, characterized in that: When normalizing the model prediction results, the geological attribute value of each grid point is converted into a probability value. After normalization, the sum of the probabilities of each grid point is 1, ensuring the comparability of geological attribute values.

7. The dynamic geological model updating and prediction method based on tunnel face information as described in claim 1, characterized in that: The 3D point cloud rendering technology is used to display dynamically updated geological models. A color mapping scheme is used to distinguish geological attribute values, and color bars are used to clearly indicate the probability distribution of geological attributes in each area. The 3D point cloud rendering uses geometric mapping technology to import the prediction results into 3D visualization software. The geological distribution is displayed intuitively through color depth, and incremental loading of dynamically updated data is supported.

Images